Subject: Re: [ecasound] Toruble with audio recording
Date: Sun Aug 10 2003 - 13:13:00 EEST
Please let me first thank all of you for your substantial help!
But unfortunately the trouble goes on. This is what I checked and did
in the meantime (the most interesting part is points 9 and 10):
1. I have OSS which comes with Knoppix Linux (a Debian system). I did
NOT switch to ALSA because of some warnings to do so.
2. I therefore have no alsamixer or other ALSA-related mixers. But I
once again checked KMix: I can mute and unmute the microphone, I
can control the volume of the microphone as well as the general
volume. Same is true for aumix (gtk).
3. I once again tried my little test recording using "ecasound
-i:/dev/dsp -o:test.wav", with the same result as before: the file
test.wav is written, but it is empty. To be precise: the file
contains this: "RIFF ...WAVEfmt ........D-...+-......data...."
(not all characters are reproduced correctly here, but that's more
or less the content of the file: 44 k with the above characters).
4. I did ecasignalview /dev/dsp test.wav and got this:
"Caught an unknown exeption"
5. When I use "play" (a sox tool) to play that test.wav-file, no error
message is displayed, i.e. the file is read, but no sound is
heard. Same is true for "ecasound test.wav". Both programs behave
as if the file exists correctly, but without a sound.
6. When I use "rec" (the sox tool) to record my test.wav, the whole
thing hangs, i.e. I cannot stop "rec" with control-c, but I have to
kill the whole application (in this case, the console). After this
there is a very high and very faint whistle (hope that's the right
word) coming from the speakers, as if there is still some process
going on. But the command "ps" shows that there is no process
hanging. Only on reboot the whistle disappears.
7. When I use "play" to play the previously with "rec" recorded
test.wav, I get this:
"Sox: Failed reading test.wav: WAVE: RIFF header not found"
8. I opened my PC and put my soundcard into another slot, just in
case... I checked all the cables; everything seems to be o.k. After
this I repeated my tests, all with the above mentioned results.
9. Then I took a look into the log-file. This might be interesting:
In "kern.log" I found:
"kernel: solo1: cannot start recording, DDMA mask bit stuck at 1"
(this message is repeated a couple four times or so). A little bit earlier
in that log-file I found:
"kernel: solo1: joystick port at 0xe001
kernel: solo1: ddma base address: 0xd800"
As far as I understand "solo1" is the sound card (isn't it?),
because in KMix the soundcard (a TerraTec 128 i PCI) is represented
as "ESS Solo1", which seems to be correct.
10. I therefore conclude that I should change the "DDMA mask bit". But I
have absolutely no idea what "DDMA mask" is and how I could change
this bit to the correct setting.
Thanks for your patience...
This archive was generated by hypermail 2b28 : Sun Aug 10 2003 - 12:58:08 EEST